Vous êtes sur la page 1sur 32

Delinquency Theories

Delinquency Theories: appraisals and applications provides a full and accessible overview
of contemporary theories of juvenile delinquency.
The book opens with a comprehensive description of what a theory is, and
explains how theories are created in the social sciences. Following on, each subsequent chapter is dedicated to describing an individual theory, broken down and
illustrated within four distinct sections. Initially, each chapter tells the tale of a
delinquent youth, and from this example a thorough review of the particular
theory and related research can be undertaken to explain the youths delinquent
behavior. The third and fourth sections of each chapter critically analyze the
theories and provide a straightforward discussion of policy implications of each,
thus encouraging readers to evaluate the usefulness of these theories and also to
consider the relationship between theory and policy.
This text is an invaluable resource for both undergraduate and graduate students of subjects such as youth justice, delinquency, social theory, and criminology.
John P. Hoffmann is Professor of Sociology at Brigham Young University. His
research interests include the etiology of juvenile delinquency and drug use over
the life course. His research has appeared in Criminology, The Journal of Quantitative
Criminology, and other journals.

Delinquency Theories
Appraisals and applications

John P. Hoffmann

First published 2011

by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Simultaneously published in the USA and Canada
by Routledge
711 Third Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
2011 John P. Hoffmann
The right of John P. Hoffmann to be identified as author of this work
has been asserted by him in accordance with sections 77 and 78 of the
Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or
utilized in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in
any information storage or retrieval system, without permission in writing
from the publishers.
Trademark notice: Product or corporate names may be trademarks or
registered trademarks, and are used only for identification and explanation
without intent to infringe.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging in Publication Data
Hoffmann, John P. ( John Patrick), 1962
Delinquency theories: appraisals and applications/by John P. Hoffmann.
p. cm.
ISBN 978-0-415-78186-2 (hardback) ISBN 978-0-415-78187-9 (pbk.) 1.
Juvenile delinquency.I. Title.
HV9069.H654 2011
ISBN: 978-0-415-78186-2 (hbk)
ISBN: 978-0-415-78187-9 (pbk)
ISBN: 978-0-203-83528-9 (ebk)
Typeset in Baskerville
by Prepress Projects Ltd, Perth, UK


List of figuresvi
List of tables

1 Theory and delinquency

2 Deterrence and delinquency


3 Biological theories of delinquency


4 Psychological theories of delinquency


5 Communities and delinquency


6 The stresses and strains of adolescence


7 Learning to be delinquent


8 Control theories of delinquency


9 Labeling, symbolic interaction, and delinquency


10 Conflict and radical theories of delinquency


11 Postmodern and feminist views of delinquency


12 Integrating and elaborating theories of delinquency


13 Developmental theories of delinquency and crime








A simple model of self-esteem and delinquency

Example of a reciprocal relationship
The basic deterrence model
Perceived risk and deterrence
Layout of street modifications from Operation Cul-de-Sac
Lombrosos classification of criminals
Sheldons three body types
Mesomorphy, temperament, and delinquency
Examples of minor physical anomalies
The DNA molecule
Neuron and neurotransmitter in the human cell
MAO-A activity, childhood maltreatment, and antisocial behavior
Normal arousal and underarousal of the autonomic
nervous system
Heart rate and criminal behavior
Major regions of the human brain
Eysencks personality dimensions
Indirect associations between intelligence and delinquency
Development pathway of CD, ODD, and other mental
health disorders
Mothers hostile attributions, discipline practices, and
childrens externalizing behavior problems
Burgesss model of concentric urban zones
A general model of social disorganization
Socialization and social control models
Parental supervision, residential instability, and delinquency
Community composition, collective efficacy, and delinquency
Mertons basic strain model
Cohens elaboration of strain theory
A final version of traditional strain theory
A test of general strain theory.
Differential association and delinquency
Delinquent friends and delinquent behavior



Social learning variables and adolescent cigarette smoking

Early-starter and late-starter models of delinquency
Role commitment, role-taking, and delinquency
Social networks: delinquent and non-delinquent
Hirschis social bond theory
A test of social bond theory
A longitudinal test of social bond theory
A test of self-control theory
Informal labeling and delinquency
Effects of police intervention on general crime
Self-esteem, delinquent peer associations, and delinquency
Power-control theory of delinquency
A test of power-control theory
A test of integrated structural-Marxist theory of delinquency
A simple end-to-end integrated theory
A simple side-by-side integrated theory
A simple cross-level integrated theory
Elliott and associates integrated model of delinquency
A test of Elliott and associates integrated model of delinquency
Social development theory of delinquency
A test of social development theory
Integrated genetic and social bond model of delinquency
Interactional theory during middle adolescence (ages 1416)
A test of interactional theory
Control balance theory
Trajectories of delinquency
Trajectories of delinquency and crime, ages 770
Interactional theory and intergenerational transmission of
antisocial behaviors
13.4 Intergenerational model of offending
13.5 Age-graded theory of informal social control
13.6 Self-control, relations with parents, and criminal behavior at
age 21




Four scenarios involving physical abuse and delinquency (percent)

Sheldons body types and associated temperaments
Lobes of the cerebral cortex
Adjectives and descriptors of Five-Factor Model personality items
Diagnostic criteria for conduct disorder and oppositional
defiant disorder
4.3 Kohlbergs stages of moral reasoning
6.1 Mertons adaptations to strain
7.1 Definitions of key concepts in social learning theory



I have been working on this book off-and-on for several years. The idea of it was
planted in my mind by Ronald Akers, distinguished criminologist, developer and
advocate of social learning theory, and author of the excellent book Criminological
Theories (co-authored with Christine Sellers). Ron was interested in a companion
book that featured theories of juvenile delinquency given that few texts focused
on this important topic. Following his invitation to prepare such a book, I ended
up thinking about it for a while and then agreeing to take on the task. However,
as often happens, the idiosyncrasies of life were at play and this project took many
more years than I had anticipated. If it were not for the support of several people
there is no doubt I would never have completed this book. First and foremost, I
thank Tim Ireland, collaborator, colleague, teacher, and friend. He helped with the
organization and writing of a couple of chapters, reviewed the entire book with
a careful and thorough eye, and helped me edit it down to a manageable length.
Karen Spence was tireless in finding various bits of information and diagrams
that helped illustrate some important issues. I thank her most enthusiastically. The
administrative staff of the Sociology Department at Brigham Young University
helped me prepare the references. Several reviewers offered many useful suggestions for which I am most appreciative. The staff at Routledge, in particular
Gerhard Boomgaarden, Jenny Dodd, and Elisabet Sinkie, as well as Andrew R.
Davidson of Prepress Projects Ltd, were incredibly supportive and helpful as they
shepherded the book through all stages of production. The final product is so
much better than I could have imagined thanks to their hard work. Finally, I thank
my family for the warm and loving support I receive every day. In particular, this
book is dedicated to my son Brian, who currently lives and serves in Ghana, of
whom I am most proud.

1 Theory and delinquency

There are many explanations of why some adolescents become involved in

juvenile delinquency. Yet, before considering any explanation, we should ask: Is
delinquent behavior simply a normal part of growing up? Studies show that about
half of adolescents in the United States are involved in some type of illegal activity
before their eighteenth birthday, so perhaps delinquency is, in some ways, normal.
However, those youths who get involved in delinquency are also more likely than
others to have school, relationship, and employment problems later in life. In addition, delinquency costs U.S. society an estimated $15billion per year. We should
therefore try to understand why some adolescents become involved in delinquency
because it may improve the lives of individuals, relationships, and society.
Given this books title, it should be clear that defining theory and delinquency
are necessary if we are to going to study the major explanations of this behavior.
In general, theories are designed to explain or predict some event or phenomenon.
Delinquency is defined as the violation of criminal laws by persons under the age of
18 (or younger, depending on the type of crime and the jurisdiction). Both of these
statements are deceptively simple and mask many underlying assumptions and
realities about the way people behave. For instance, many theories of behavior,
whether they involve delinquency or musical tastes, make reasonable claims, yet
fail when used to explain the way people actually behave. In addition, the term
delinquency, similar to the term crime, is often not useful. Buying cigarettes one time
and killing multiple people both fall under the umbrella of delinquency if committed by a 13-year-old, but should we treat them the same way? Can we explain
both behaviors using a single theory? Delinquent behaviors are extremely diverse
and it is best to admit that understanding them with theories is a complex exercise.
Before studying theories of delinquency, it is therefore necessary to consider some
of these concerns.

What is a theory?
As mentioned earlier, a simple definition of a theory is that it is an attempt to explain
or predict some event or phenomenon. Many of us claim to have theories of why
something occurs. However, we often confuse an opinion with a theory. Opinions
are judgments or beliefs that are not necessarily based on facts or generalized ideas

2 Theory and delinquency

about truth. I may have the opinion that American Idol is the best reality show on
television, but others may prefer The Amazing Race. My opinion is not a fact, nor
is it a theory. There is also some confusion about the differences between facts
and theories (Sears, 2005). Facts are things that have actually occurred; they are
particular truths known by valid observation. I know for a fact that my child was
born in May 2000 because I was there to witness his birth. Of course, facts may
be disputed: I claim that Michael was smoking cigarettes in the school parking lot,
whereas Jennica challenges this because she was in geometry class with Michael
during this time. One of us is stating an actual fact; both observations cannot be
Theories are more like models: They provide a picture or general representation of how a phenomenon has occurred. Suppose, for instance, that Im driving
home along a deserted stretch of highway and see a pool of water on the road up
ahead. I havent seen any clouds in the sky, nor have weather reports mentioned
rain. The next day and each day thereafter I see the same pool of water, yet there
has been no reported rainfall. The curious thing is that when I drive past the place
where it appears the pool should be, its not there. So I think I must be seeing a real
pool of water that somehow disappears. Perhaps there is an underground spring
that bubbles up, but then the water drains quickly into a hole by the time I reach
the spot. This is a simple explanation of why I see the pool of water each day, but
I could make it more complex by studying the size of the pool, recording the time
of day, scanning weather reports, and considering the movement of water on hard
The fact that I developed a theory explaining the disappearing pool of water
doesnt make it accurate. Most people who have experienced seeing water on
a stretch of highway know that it is an illusion caused by the way the suns rays
reflect off the dark surface of a road. Nevertheless, my attempts to explain this
phenomenon are not unlike the way that people develop theories of all kinds. For
example, why do some migratory birds fly in a V formation; why do stars appear
to move across the night sky? The more curious among us have tried to explain
why these events occurred in other words, they developed theories to explain an
event or phenomenon.
The desire to come up with theories is a normal part of human existence.
Some have also studied the process of creating theories. For example, here are
some formal definitions of theory used in the social sciences.
A theory is an attempt to explain something or describe the causes of something (Agnew, 2008, p.72).
A theory is a set of logically interrelated statements in the form of empirical
assertions about properties of infinite classes of events or things (Gibbs, 1972,
Theory is a mental activity; ... it is a process of developing ideas that can
allow us to explain how and why events occur (Turner, 2003, p.4).

Theory and delinquency3

Theory: A scheme or system of ideas or statements held as an explanation or
account of a group of facts or phenomena (Oxford English Dictionary, 2010).
Although defining theory is a controversial topic in the social sciences (Sica,
1998), these four definitions provide a general sense of how this term is defined.
All four note that theories are designed to explain or make sense of some event
or phenomenon, much as I tried to do when describing the pool of water. Turner
(2003) furthermore defines theory as a process, thus reminding us that good
explanations require several steps. Many social scientists also prefer to limit their
theories to scientific theories, or those that can be tested using scientific methods.
Therefore, designing theories falls under the general field of science.
But what is science? In general, science is characterized by the idea that we may
develop knowledge about how the world actually works, and this knowledge may
be altered by carefully observing empirical (measurable) events. For instance, suppose my favorite television program is Cops. Ive noticed that many of the people
who are arrested on Cops are young men without shirts. I develop the idea that
being male and not wearing a shirt are pieces of knowledge that are tied to being
arrested by the police. If so inclined I can represent this idea using the following
simple equation:
male+no shirt=arrest by police.
I can then test my crude theory of police arrest by observing not only the television show Cops, but also other reality TV shows, talking to the police about people
theyve arrested, asking those I know who have been arrested whether they are
male and wear shirts, and discussing my theory with friends and experts on policing. I might even play the role of a scientific researcher by developing an intricate
study of police arrests, complete with a sample of arrestees, a team of expert
observers, and statistical techniques to analyze the data these observers collect. In
this case, my theory will be proven wrong by testing it against actual events, even
though I may not directly observe the sample of arrests. Some people argue that
being proven wrong is the key for advancing science and for developing better
theories (Popper, 1968): Once we eliminate theories that claim that shirtless young
men get arrested, we may focus on other, more important aspects of people who
get arrested and develop better theories about them. This approach is known as
falsification. However, I do not have to discard all the parts of my theory. I will
probably find, for example, that males are arrested more often than females, so I
should try to develop a better theory that applies to this piece of knowledge.
There are numerous theories of human behavior that have been developed
in many different academic disciplines. In general, regardless of the discipline,
theories should consist of concepts and statements. Concepts are the principal elements of a theory and they provide labels for specific phenomena. Statements,
which are sometimes called propositions, link the concepts together and indicate
why concepts are connected. An example of a concept is self-esteem, which is
defined as the beliefs or feelings people have about themselves. Another example

4 Theory and delinquency

of a concept is delinquency, which was defined earlier in the chapter. A theorist who
wishes to develop a theory of self-esteem and delinquency must figure out a way
to link these two concepts with a statement or proposition. Often, theories are
represented visually by graphical models. Figure 1.1 provides a simple model of
self-esteem and delinquent behavior. The two concepts are represented as ovals.
The statement is represented by the solid arrow connecting the two concepts.
We can determine two things about the proposed link between self-esteem and
delinquency from the figure. First, the theorist thinks that self-esteem leads to
delinquent behavior. If the theorist thought that delinquency affected self-esteem,
the arrow would point in the other direction. Second, notice the negative sign
above the arrow. This suggests that, as self-esteem decreases, delinquent behavior
increases. The next step for the theorist is to describe why these two concepts are
linked. A figure cannot fully represent a theory; the theorist usually must also use
words (or equations) to describe the links.
How we define concepts often causes a lot of confusion in studies of delinquency. For instance, is self-esteem an easy concept to understand; does the concept
mean the same thing to everyone? What about the concept of delinquency; does
it mean the same thing in each theory? As you review the theories of delinquency
presented in the following chapters, pay attention to how the main concepts of the
theories are used.

What makes a good theory?

A good theory should provide clearly defined concepts and statements linking
these concepts. A good theory should also offer a compelling explanation of whatever phenomenon it is concerned with. However, how do we know if it provides a
compelling explanation? Most observers contend that a good theory should offer
the following characteristics: coherence, verifiability, simplicity, significance, scope,
and utility (Mithaug, 2000).
Coherence refers to whether the theorys statements are logically connected in a way
that offers a clear explanation of the phenomenon. The example of the shirtless
young men who are arrested on Cops does not pass the coherence test. It may be
coherent in the sense that most people can visualize the concepts of shirtlessness


Statement (proposition)
Figure 1.1 A simple model of self-esteem and delinquency.


Theory and delinquency5

and arrest by the police, but the statement linking these two concepts offers only a
vague explanation of why they are connected. Is there something about shirtlessness that makes young men violent or prone to commit crime? This seems highly
unlikely. Do the police target these young men for arrest? Perhaps, but explaining
why is difficult.1 Therefore, unless the statements in a theory are coherent in light
of common sense and reasonableness, the theory fails.
Another aspect of coherence is whether the theory is non-tautological. In the
field of logic, a tautology is a compound proposition that is true for all possibilities of
its elementary propositions (criminology is what criminologists do). In practical
terms, a tautology is a theoretical statement that is always true no matter how the
outcome is considered. For example, if my theory claims that delinquent behavior
results from the presence of delinquent values among youths, I run the risk of tautological thinking. If Im not cautious, I could be asserting that delinquency results
from delinquency: What is a delinquent? Someone who has delinquent values.
And who has delinquent values? Delinquents. However, if I carefully define delinquent behavior and delinquent values as distinct phenomena (see Chapters 6 and
8) and then link them in a coherent way, I can avoid a tautology.
Verifiability refers to whether the theory is supported by empirical evidence or measurable results. This is often called empirical validity. A theory is verifiable if research
supports its statements. Tests designed to verify theories (or challenge them) should
involve research that is well thought out and conducted in a rigorous manner. One
of the problems faced by researchers is that measuring social phenomena, such as
values, self-esteem, and delinquency, is not an easy task. Most books on research
methods discuss in detail how challenging it is to collect good information from
people about their behaviors, attitudes, and emotions.
Another key issue for theories is whether they should explain or predict all
cases of some phenomenon, or simply provide better explanations or predictions
than other theories. Should the statements of a good theory of delinquency
always be true? When we say that some concept such as weak emotional attachment to parents causes delinquency, must we always find this link? There is a
large literature in the philosophy of science and the social sciences on the issue
of causality (e.g., Pearl, 2000), with many debates about whether we can ever
establish that one concept causes another when they involve human behavior.
Perhaps the most well-known, but still controversial, description of causation is
by the English philosopher John Stuart Mill (1884). He defined causation based
on three criteria:
1 The cause precedes the effect in time.
2 The cause and the effect are associated: As one changes the other also changes.
3 Alternative explanations of the relationship between the cause and the effect
can be eliminated.

6 Theory and delinquency

Some observers have interpreted these criteria as demanding that, for one thing
to cause another, the effect must always change with the cause. Others argue that
we simply must establish that the cause increases the likelihood of the effect. Most
important, we wish to eliminate alternative explanations for the link between the
cause and the effect. It is often the case that a third factor causes both.
As an example, suppose my theory predicts that physical abuse by parents causes
their children to engage in delinquent behavior. A strict view of causality claims
that my theory is correct only if (1) physical abuse comes before delinquency;
(2) as physical abuse increases, delinquency also increases; and (3) some other
factor does not explain this association (e.g., exposure to lead in the familys diet
does not cause both). However, is my theory incorrect if someone finds that some
abused children are not delinquent? Most observers now see a good theory as one
that offers a good probabilistic explanation of the phenomenon. Some researchers
simply avoid using the term causality to avoid the debate over its meaning. In this
book, the term causal is not used unless a particular theorist uses it. Rather, using
the example of physical abuse, we adopt the following position: A theory that
states that parental physical abuse leads to delinquency is not necessarily incorrect
if we find a group of abused children who do not engage in delinquent behavior.
The important point is whether children who are abused are more likely to be
(have a higher probability of being) delinquent than those who are not abused.
Of course, we must now make some judgment calls about how much more likely
they must be before we consider the theory to be verified. This complicates things
because I might be willing to accept one probability as sufficient, whereas another
person might want the probability to be higher.
Table 1.1 illustrates four scenarios involving parental physical abuse and
delinquency. The first scenario supports a deterministic theory: All children who are
abused are delinquent, whereas none who are not abused are delinquent.2 The
next set of scenarios show three probabilistic situations. In the second scenario 5
percent of abused children are delinquent, whereas only 1 percent of non-abused
children are delinquent. In the third scenario 25 percent of abused children are
delinquent and 15 percent of non-abused children are delinquent. In the fourth

Table 1.1 Four scenarios involving physical abuse and delinquency (%)
Scenario 1
Scenario 2
Scenario 3
Scenario 4


No delinquency

Physical abuse


No physical abuse


Physical abuse


No physical abuse


Physical abuse



No physical abuse



Physical abuse



No physical abuse



Theory and delinquency7

scenario, 75 percent of abused children are delinquent and only 10 percent of
non-abused children are delinquent. Which piece of evidence, if any, verifies
the theory that physically abused children are more likely to be delinquent? A
strict determinist would claim that only the first scenario verifies the theory. One
observer might claim that the second scenario is persuasive: Abused children are
five times as likely as non-abused children to be delinquent. A second observer
might contend that Scenario 4 provides the best evidence: Three-quarters of
abused children are delinquent, but only one-in-ten children who have not been
abused are delinquent. This shows the blessing and curse of probabilistic thinking when addressing theories: Asking our theories to predict probabilistically is
a better reflection of the complexities of human behavior, yet there are no fixed
rules for deciding when a probability is high enough to satisfy our desire to verify
a theory.
However, there is another issue that often complicates tests of delinquency
theories and has forced some researchers to rethink the way that the delinquency
process is envisioned. Although we usually assume that some relationship or
behavior leads to delinquency, it is often the case that delinquency affects one of
these. For example, learning theories (see Chapter 7) propose that youth learn
to be delinquent from their friends, thus delinquent friends lead to delinquency.
Research shows, however, that delinquent behavior also leads to associating with
delinquent friends. This is an example of a reciprocal relationship (Figure 1.2): Factor
1 leads to factor 2, but factor 2 also leads to factor 1. In this reciprocal loop, each
factor may be considered a cause of the other.
Relying only on empirical predictions for verification also asks too little of these
theories. Two distinct theories may offer similar predictions about who becomes
delinquent. Twenty-five percent of abused children may become delinquent, but
suppose that we also find that 25 percent of children with low self-esteem become
delinquent, thus supporting a theory that links self-esteem and delinquency. How
do we decide which theory is better? The better theory will offer a better explanation
for delinquency, not simply a more precise prediction. In fact, good explanations
of delinquent behavior will improve predictions of delinquent behavior (Douglas,
2009). Moreover, we should be prepared to modify our theories when faced with


Figure 1.2 Example of a reciprocal relationship.

8 Theory and delinquency

empirical evidence. In the preceding example, perhaps abuse and low self-esteem
are part of a poor family environment, which then increases the probability of
delinquency. In general, verification is an essential characteristic of a good theory,
but it must be accompanied by other important elements.
Good theories should be as simple as possible, without ignoring the complexities
of human behavior (Lave and March, 1975). There are many theories that contain
well-defined concepts and persuasive statements, but they lack simplicity because
they include concepts or statements that are unnecessary. For example, theories
that attempt to account for every conceivable relationship often lack simplicity. It
is better to look for a key set of concepts and statements that explain the phenomenon rather than taking an overly comprehensive approach to theory building.
A theory that explains delinquency based on differences in levels of self-esteem
is more parsimonious than a theory that explains delinquency based on differences in levels of parental attachment, blood sugar, language skills, fiber intake,
and self-esteem. Nonetheless, a theory that meets the simplicity goal is not necessarily valid if, say, self-esteem offers a poor explanation for delinquency. In the
quest for simplicity, verifiability, coherence, and the other characteristics of a good
theory should never be sacrificed.
Significance, scope, and utility
The next three characteristics of a good theory significance, scope, and utility are concerned with whether a theory has value. A theory may be highly
predictive, coherent, and parsimonious, but may be useless if it addresses a very
narrow phenomenon or one that has little concern for society. A theory about
why 20-year-old women are more likely than 18-year-old men to play the bass in
rock bands may be coherent and verifiable, but that doesnt make it valuable. A
significant theory should provide an explanation for all (or at least most) of the
key factors that lead to the phenomenon of interest. Are there factors that are well
known to predict the phenomenon, yet are left out of the theory? An example
from delinquency research involves friends. As mentioned earlier, a consistent
empirical finding is that adolescents involved in delinquent behavior are more
likely than others to have friends involved in delinquent behavior. A good theory
should at least consider the role of friends in its explanation of delinquency. If it
does not, then we should question the significance of the theory.
A theorys scope concerns the range of phenomena it explains. Some social
scientists argue that social theories should be general enough to encompass
numerous human behaviors, from normal to deviant. When assessing delinquency
theories, it is a good idea to consider the following question: If the theory does a
fine job of explaining delinquency, does it also explain law-abiding behavior? Is it
useful for explaining various types of behavior, from shoplifting to auto theft? One
of the problems for these theories involves their scope: Delinquency covers a lot

Theory and delinquency9

of behavioral ground. Is it asking too much to expect them to explain all types of
delinquency? Alternatively, should we strive for multiple theories of delinquency
that explain various behaviors, such as running away from home, drug use, and
violence? It should be clear that there is a delicate balance between scope and
simplicity: Although we should strive for simplicity, we must not ignore scope.
A theory may be quite parsimonious, with only one or two concepts linked to
delinquency. However, if its scope is too narrow, then it risks the curse of having
little value. On the other hand, it is not a good idea to develop theories that try
to encompass every social phenomenon and then attempt to include every concept that has been linked to these phenomena. Overly broad theories can also be
Finally, an important question is whether a theory has utility: Does it offer
practical solutions to the problem of interest? Can it be used by policy makers?
Alternatively, is it so obscure that it is almost impossible to determine what to do
with it? A theorys utility is based on all of the other factors: coherence, validity,
simplicity, significance, and scope. Public policies must be sensitive not only to
whether programs follow good theories, but also to limited resources, political
needs, and bureaucratic realities. It is unlikely that a program will be implemented
if it is based on a theory of juvenile ant farmers and prostitution, no matter how
good the theory. This type of theory is simply too narrow in scope and has little
significance to broader issues of delinquency to provide much utility.

Classifying theories of delinquency

There are several classification schemes for theories of behavior. The most basic
scheme when addressing delinquency is whether we are concerned with how
juveniles behave or with how the public and government agencies react to how
they behave. The first topic how juveniles behave has led to many theories
of delinquency. These will be emphasized throughout this book since they are
the most common and have generated the most research. However, we should
also remember that the study of delinquency is driven by how delinquency is
defined. There are some who argue that delinquency is defined arbitrarily, that
those concerned with explaining adolescent behavior should not limit themselves
to behaviors that are defined by the criminal justice system and by legislators who
create laws. Yet the most common definition of delinquency is that it involves
behaviors by persons under the age of 18 (or younger, depending on the type
of crime and the jurisdiction) that violate the criminal laws. The boundaries of
delinquency are created by lawmakers and may even shift depending on differing
state or local laws. A good example involves curfew laws, which make it unlawful
for young people to be in public places during certain hours, usually at night. They
are an example of a status offense: acts that are not legal for persons under a certain
age, but that are legal for adults. This may complicate our desire to explain delinquency because status offenses treat particular behaviors differently if committed
by adolescents or by adults. Moreover, because the definition of a status offense
often changes from one area to another, one might define a group of young people

10 Theory and delinquency

as delinquent if they violate their citys curfew law, even though they would not be
violating the law if they lived in a city without a curfew law.
Some researchers prefer to study not the violation of law by juveniles, but
rather the creation and enforcement of these laws. This is an important topic
that has led to many fascinating studies. Two key questions are (1) why do legislators pass certain laws that are targeted at the behaviors of juveniles and (2) are
the enforcement efforts of governmental agencies more likely to be focused on a
particular set of juveniles, such as minority youths?
It is wise, then, to remember that many studies of delinquency use arrest or
police contact information as the basis for defining this behavior, yet these may
be affected by differential enforcement of laws. Official bias the tendency to disproportionately arrest or institutionalize certain adolescents based on something
other than their law-violating behavior is an important factor in delinquency
research that cannot be ignored. As you consider the research that tests theories
of delinquency, consider also the source of data used to define delinquency. Could
it be biased against certain groups of adolescents? Are theories that try to explain
differences between groups of adolescents (e.g., males vs. females) tested using
data that might also be biased? It is essential that we pay attention to the interplay
between theory and research in studies of delinquency. Many of the most prominent theories are based on studies of juvenile behavior; most have undergone
revisions because of the results of these studies.
Whose behavior is being explained? Individual-level vs.
macro-level theories
Theories of delinquency may be classified into those that address individual-level
behavior or those that address macro-level behavior. Individual-level theories
seek to explain behaviors by individuals and the factors that affect them. Most of
the examples provided earlier are drawn from individual-level theories: Physical
abuse leads to delinquency; delinquent friends lead to delinquency. In contrast,
macro-level theories attempt to explain delinquency across spatial units, such as
neighborhoods or cities. A simple macro-level theory contends, for instance, that
neighborhoods higher in residential instability (residents move in and out at a
relatively rapid rate) also have higher rates of delinquency due to characteristics
of these neighborhoods (see Chapter5).
Attention to individual-level and macro-level phenomena does not exhaust the
ways we can view delinquency theories. In between these two levels are theories
of small group behavior. Research on gangs and delinquent peers often uses these
theories to understand how groups behave. Since delinquency is frequently a small
group activity groups of youths fight, friends go on a shoplifting spree or steal
a car considering how small group dynamics affect adolescent behaviors is a
useful method for understanding delinquency (e.g., Krohn, 1986). As discussed
in Chapter 7, social network theories, in particular, have been used to explain the
group dynamics that affect delinquency.

Theory and delinquency11

Assumptions about human nature
A third way to classify delinquency theories is based on assumptions about human
nature. For example, are people naturally inclined to act in the best interests of
their fellow human beings? Alternatively, are people naturally selfish and apt
only to look out for their own interests unless we design a system to curb these
impulses? Do people emerge from the womb as blank slates who must be taught
everything about how to behave, or are there certain parts of our minds that are
fixed before birth and affect how we behave? Do political, economic, and social
systems influence the way we act? Is conflict a healthy part of human existence as
different groups seek to advance their interests, or should we strive to avoid conflict? Seeking answers to these questions has occupied the minds of great thinkers
for centuries and has led to many ideas about human behavior.
In studies of delinquent behavior, answers to these questions have led to a
general distinction between consensus and conflict explanations of crime and delinquency (Bernard, 1983). Consensus theorists view society as, at best, cohesive and
made up of individuals who follow or wish to follow the accepted rules of
behavior. The preferred society is made up of people who agree about proper
forms of behavior and support a particular moral order. When behavior falls
outside the proper boundaries established by society it must be eliminated or else
society is damaged. Most delinquency theories that rely on legal definitions of
unlawful behavior follow a consensus model.
Conflict theorists view the typical society not as a cohesive group designed to
curb or encourage certain behaviors, but rather as consisting of separate groups
striving to gain, maintain, or manipulate power. A moderate form views conflict as
a necessary condition in the production of society. Since most people are naturally
drawn to group associations, there tend to be groups in any society that have
conflicting interests. The conflict among these groups typically leads to stability. In
fact, resolving conflict is a normal way that societies develop.
Delinquency theorists who take a conflict approach admit that certain juveniles
are more involved in misbehavior, but then ask what it is about their position in
society that affects their behaviors. Perhaps their parents work in lower-wage jobs
that provide little autonomy and emphasize authoritarian styles of interpersonal
relationships. These parents then treat their children poorly or do not teach them
to view other people kindly. The result is a higher likelihood of delinquency among
juveniles from lower-status or authoritarian families (see Chapter10).
The consensusconflict divide offers only a broad view of how to classify theories of delinquency. A more detailed classification scheme is based on whether
we view human nature as being naturally selfish, selfless, or a blank slate. The
philosophical view that human nature is selfish or that self-interest is part of
human nature finds its origins in the British philosopher Thomas Hobbess (1962
[1651], p.113) argument that the scarcity of valued things in the world leads to a
constant war of all against all and the saying that life in the primordial state of
nature is solitary, poor, nasty, brutish, and short. We require a social order that
includes fixed laws or else people will destroy each other. This assumption about

12 Theory and delinquency

human nature leads to theories that ask not why people violate laws, but rather:
Why do [people] obey the rules of society? (Hirschi, 1969, p.10). Since people
are naturally inclined toward selfishness and exhibit behaviors that result from this
tendency, theories of delinquency should describe the factors that restrain juveniles from committing crime, rather than what factors compel them to commit
crime (see Chapter8).
Suppose we assume, as did the French philosopher Jean-Jacques Rousseau
(1974 [1755]), that people are naturally good (noble savages) and are corrupted only by societys less noble pursuits. We should then assume that people,
if left on their own, would choose to help other people and act with selflessness.
Unfortunately, the formation of governments often allows certain groups to gain
too much authority over the behavior of others. What might a theory that accepts
this assumption say about delinquent behavior? It would probably look to the goals
people are encouraged to have and also at the means they have for achieving these
goals. Some groups, especially those with less power, have fewer means than others
to achieve goals, so the theorist would look for ways that they compensate for their
social disadvantage (Merton, 1968). Some may turn to crime and delinquency to
reach these goals when legitimate avenues are blocked. As we shall see in Chapter
6, juveniles are a group in our society with less power than others and are often
not given direct access to certain valuable goals, such as owning a car. They are
pushed into delinquency, rather than acting based on their natural impulses.
Another view of human nature is that our minds at birth are a blank slate
and we must have experiences to learn to function in society. The British philosopher John Locke (1975 [1690]) believed that all our ideas and knowledge come
from experience. We have sensations that teach us about what occurs in the world
external to us, and then we reflect on these sensations to learn how we should use
our own abilities to operate in the world. Although contemporary studies show
that we learn through a more intricate system than Locke envisioned, and that we
are not blank slates at birth (Pinker, 2002), the idea that we must learn to function in the world through our experiences has led to a denial that human nature is
naturally good or selfish. Rather, we have individual natures that depend on what
we have learned throughout childhood and into adulthood. Theorists who accept
this assumption about human nature have developed several learning theories of
delinquency (see Chapter7).
Although these three assumptions of human nature are broad enough to
encompass most theories of delinquency, there are also other assumptions that do
not fit neatly under this classification scheme. For example, the British philosopher
Jeremy Bentham (1996 [1789]) regarded the main drive behind human behavior
as the pursuit of pleasure and the avoidance of pain. The father of economics,
Adam Smith (1991 [1776]), believed that the origin of human behavior lay in the
pursuit of self-interest. Although this is similar to Hobbess position, Smith also
argued that the accumulation of individuals self-interested pursuits would be the
best economic outcome for the most people. By meeting others self-interest with
our own, we may gain the greatest benefit for all. Contemporary rational choice
theorists have suggested that people behave rationally when they select the best

Theory and delinquency13

means available to reach a goal they think is in their interest. The assumptions
promulgated by Bentham and Smith have led to several theories of delinquency
that emphasize deterrence and rational choice (see Chapter 2). If people pursue
their own self-interests in a rational manner and these interests involve the maximization of pleasure and the minimization of pain, then we should consider that
some people may commit delinquent acts as a way either to gain things they view as
pleasurable (e.g., money) or to avoid pain (e.g., escaping punishment). Deterrence
theorists argue that we must develop ways to make delinquency more painful (or
risky) than pleasurable. This might be done through having punishments whose
pain (not just physical pain, but other types such as deprivation of desired
goods) exceeds the pleasure derived from delinquent behavior. Alternatively, it
might occur as policy makers take steps to make delinquency more risky, thus
reducing the rewards of misbehaviors.
A final assumption about human nature that has led to several theories of
delinquency does not focus on people in some hypothetical state of nature. Instead,
theorists such as Charles Horton Cooley (1902) and George Herbert Mead (1934)
considered the development of ones view of oneself and others through social
interactions. Although their ideas are not entirely distinct from learning theories,
Cooley and Mead were more concerned with the development of individuals
self-image and how this emerges through the many interactions they have with
others. Mead furthermore thought that, because people cannot do all the things
they need to to survive by themselves, there is a natural tendency to rely on group
participation and cooperation to survive. Therefore, individuals views of themselves and the roles they play in society depend on how others see them and how
they are treated. These assumptions about the role that interactions play in defining us have resulted in theories about symbolic interaction. These theories emphasize
that how people perceive reality is built through their daily interactions with other
people and with groups. They also address how the symbols of everyday life, such
as signs, gestures, and language, affect interactions among individuals and groups
(see Chapter9).
Finally, symbolic interaction approaches tend to consider two additional
aspects of human behavior: agency and emotions. Agency is purposeful action
on the part of individuals; we are not passive observers who are pushed around
indiscriminately by social and personal forces outside of our control. Rather, we
have the capacity to act independently, making our own choices in many situations. Emotions are spontaneous feelings that occur as we are exposed to different
situations; they are often seen as the antithesis of reason. Both of these have been
largely neglected by many delinquency theories, such as social control or social
learning (Giordano, 2010; see Chapters 7, 8, and9).3

Theory and context

We have now learned some useful background material about theory and how
to classify explanations of delinquency. In addition, it is important to realize that
theories of delinquency are influenced by the social context that theorists are

14 Theory and delinquency

part of (Lilly et al., 1994). Our ideas about the natural and the social world are
influenced in large degree by current events and shifts in scientific knowledge.
Adam Smith, the father of economics, developed his ideas during a time when
there was a growing belief in England and in the fledgling American colonies
that free trade between nations was vital (Ross, 1995). Therefore, it should not
be surprising that he emphasized that, if left on their own to pursue economic
gain, the most people would enjoy the greatest benefit. Karl Marx, one of the
founders of conflict theory, wrote during a period when capitalism was in overdrive in Europe and many factory workers lived in harsh conditions. His theory
that capitalism influenced human nature to emphasize selfishness, but that this
would change if the economic system became more egalitarian, was based on
observations about social conditions during his lifetime. During the 1960s delinquency theorists focused on economic and social barriers that affected the poor
and minorities, so many explanations emphasized how poverty and discrimination
led to delinquency.
Thus, it is important to remember that theories are tied to historical conditions.
We cannot predict which theories of delinquency will be most widely accepted
100 years from now, but there is little doubt that they will be influenced by the
social conditions and ideologies prevalent at that time. This does not mean that
we should not trust current theories of delinquency because we should anticipate
that they will change with shifting social conditions. Rather, it is wise to consider
that explanations of delinquency do not emerge in isolation from broader societal
conditions. As you learn about the theories presented in this book, it is useful to
use your knowledge of history to increase your understanding of how they may
have developed. It is also important that you consider the social context that you
grew up in and how this affects your beliefs, opinions, biases, and theories about

The goal of this chapter has been to present some important background material
that will assist you as you learn about specific theories of delinquency. We have
considered definitions of theory, in particular scientific theory, and some of the
characteristics that a good theory should possess. A good theory should be coherent, verifiable, simple (but not overly so), significant, and have sufficient scope and
utility. There are many policies and programs designed to minimize delinquency,
so it is important that we seek good theories to guide these policies and programs.
This chapter also considered several ways of classifying theories of delinquency. The key ways are by placing them into consensus or conflict categories;
categories based on whether the theories address macro-level rates of delinquency,
individual-level delinquent behaviors, or small group behaviors; and categories
based on our assumptions about human nature (Is it naturally selfish, selfless, or
a blank slate? Do humans act rationally or do emotions drive behaviors?). All of
these schemes are potentially beneficial, since each considers important assumptions about the origins and motivations of human behavior.

Theory and delinquency15

Finally, you should consider the historical circumstances that affected the
emergence of these theories. Look beyond this book to your own understanding
of the social conditions that were present at the time these theories emerged. In
addition, adopt a critical eye when considering theories of delinquency. Is the
scientific method the best way to understand and explain delinquent behavior?
Or are there other methods that are more useful or that overcome some of the
limitations of the scientific method? By considering the social context under which
these theories developed, and the methods used to examine them, we may develop
a better understanding of their promise for explaining delinquent behavior.

2 Deterrence and delinquency

This chapter is concerned with one of the oldest theories of crime and delinquency:
deterrence theory. A key assumption of this theory is that people, both young and
old, make decisions about how to behave based on a rational calculation of its
risks and rewards. This may sound simple, but deterrence involves a much more
complex set of circumstances than one might think. Consider two U.S. Supreme
Court decisions regarding the death penalty for those who commit capital crimes as
juveniles Stanford v. Kentucky (1989) and Roper v. Simmons (2005). The facts in Stanford
and Roper are quite similar. An adolescent committed a murder, showed little or no
remorse, there were aggravating circumstances (such as robbery, sexual assault, or
kidnapping), and the trial court sentenced the convicted adolescent to death.1
Prior to Roper the leading death penalty case for juveniles was the Stanford case,
decided just 16 years earlier. The court in Stanford held that the execution of an
offender convicted of a capital crime, who was either 16 or 17 at the time of the
offense, is allowed by the U.S. Constitution. Although Stanford was a controversial
decision, it was the law of the land long enough for 19 juvenile offenders to be
convicted and executed for capital crimes (Amnesty International USA, 2005).
One of the arguments raised by the defense in the Stanford case was that the
death penalty for juveniles fails to serve the legitimate goal of penology ... it fails
to deter because juveniles, possessing less developed cognitive skills than adults,
are less likely to fear death. The Supreme Court rejected this argument, claiming
as the socio-scientific data suggests ... it is not demonstrable that no 16-year-old
is adequately responsible or significantly deterred. It is rational, even if mistaken,
to think the contrary. Therefore, the Supreme Court in 1989 suggested that if the
threat of death deterred even one 16-year-old from committing a capital offense,
then a legitimate societal goal is achieved.
In the Roper case the U.S. Supreme Court, by a vote of five to four, broke
with precedent and determined that the execution of juvenile offenders who committed capital crimes was unconstitutional. A particularly important part of its
decision stated:
[a]s for deterrence, it is unclear whether the death penalty has a significant or
even measurable deterrent effect on juveniles ... [T]he absence of evidence
of a deterrent effect is of special concern because the same characteristics
that render juveniles less culpable than adults suggest as well that juveniles

Deterrence and delinquency17

will be less susceptible to deterrence. In particular, ... [t]he likelihood that
the teenage offender has made the kind of costbenefit analysis that attaches
any weight to the possibility of execution is so remote as to be virtually
nonexistent. To the extent the juvenile death penalty might have a residual
deterrent effect, it is worth noting that the punishment of life imprisonment
without the possibility of parole is itself a severe sanction, in particular for a
young person (Roper v. Simmons, 2005, pp.1718).
The reasoning of the Supreme Court in the Roper case was different from in the
Stanford decision on the issue of deterrence and how it might affect adolescent
decision-making. So which strategy is correct? Can juveniles be deterred from
crime or, because of their status as juveniles, is deterrence largely ineffective?
The U.S. legal system relies mainly on the idea that people are rational decision
makers. If people are to obey the law, then there must be punishments that outweigh the rewards for breaking the law. This type of thinking also lies at the heart
of contemporary theories of rational choice and deterrence, although we shall
learn that there are other aspects as well. However, what is meant by deterrence,
and what is deterrence theory? Furthermore, how could the Stanford court readily accept the role of deterrence in juvenile capital cases, whereas the Roper court
clearly doubted that deterrence is effective for juveniles who commit murder?
To address these questions, as well as related theoretical issues, this chapter outlines the basic arguments of deterrence theory. We begin with a discussion of the
origins of deterrence theory. We then discuss some of the major complexities of
deterrence theory, especially how punishment a core component of the theory
is presumably most effective when it is swift, certain, and severe. We also describe
the various forms of deterrence that are thought to affect decision-making among
juvenile offenders. A key part of this chapter involves assessing the research on
the effectiveness of deterrence: Does it prevent or minimize delinquent behavior?
We conclude by evaluating the promises, problems, and policy implications of
deterrence theory.

The evolution of ideas about deterrence and

The starting point for an introduction to deterrence theory is a short book published in 1764 by the Italian philosopher Cesare Beccaria entitled On Crimes and
Punishment. His work helped lay the groundwork for deterrence theory as a crime
control strategy. At the time of its publication the system of justice was quite
different from today. Importantly, there was not a separate juvenile justice system,
nor was there a distinct notion of juvenile delinquency that was separate from the
concept of a criminal. Therefore, Beccarias work applied to the criminal offender
in general, with little regard for the age of the offender.
Over time, as the concept of a separate period of life known as adolescence
was recognized, a distinction between juvenile and adult offenders emerged.
Moreover, during the late nineteenth century, one of the first formally defined juvenile courts appeared in Chicago. From this Chicago court, the idea of a juvenile

18 Deterrence and delinquency

justice system, separate and distinct from the adult criminal justice system, spread
across the country and today remains an integral part of the justice system in the
United States. Since the inception of the juvenile court there remains an unresolved debate over deterrence as a crime control strategy for delinquent behavior,
however. Even though deterrence theory continues to be the basis for much of the
adult criminal justice system, those who organize and study the juvenile system
have remained skeptical about its place in evaluating juvenile offenders. In fact,
much of the historical emphasis in the juvenile justice arena has been on treatment and rehabilitation rather than on deterrence through punishment.

The foundations of deterrence theory

The rational person
Beccaria saw as the heart of deterrence theory the assumption of a rational
person who makes choices and reaches decisions based upon a calculation of
risk and reward, punishment and pleasure. It is the game that gamblers play on a
regular basis: How much risk is involved? How much is the potential reward? If
the rewards outweigh the risks, then the person is likely to choose to behave in a
certain way. A more general area of research that addresses these issues involves
rational choice theory (Matsueda et al., 2006).2
Clearly, deterrence theory is strongly grounded in choice, and the choice is
considered rational and reasonable to the decision maker based upon all available information at the time, or what some economists call the information set. The
fundamental premise of deterrence theory is that we all choose; that the choices
are rational based upon our interpretation of pleasure and pain generated by the
situation we are confronted with and the information available to us; and that the
decisions are ours to make. Consistent with accepted principles of the criminal
and juvenile law, because actors have decided to commit acts they are blameworthy or culpable and deserve punishment.
However, theorists have also noted that the way people judge the costs and
benefits of their behaviors is affected by their past experiences and the experiences
of others. Suppose that a young person has been caught in the past for trying to
shoplift. This may change her calculation of future chances of getting caught
shoplifting. Now, suppose she has successfully shoplifted 10 times in the past week.
This may create a sense that there is little risk to shoplifting. Similarly, if her two
best friends have shoplifted without getting caught, then they may encourage her
that the risks are low and the rewards are high. In general, this means that ones
judgment of the rewards and risks of delinquent behavior can change over time
and is affected by experiences and by ones peers (Matsueda et al., 2006).
Deterrence and punishment
According to deterrence theory the way to prevent future delinquent acts is to
adjust the risk (pain) of the act so it outweighs its rewards (pleasure). Generally,

Deterrence and delinquency19

deterrence theory has focused on punishment as a way to adjust risk. As the risk
of punishment increases to surpass the reward of committing the act, a rational
individual will choose not to violate the law. It is through this mechanism of deterrence that delinquency is prevented. In brief, a person is deterred from committing
the offense because the risks (Ill get caught and punished) outweigh the rewards
(Ill get money; Ill get back at an enemy).
However, what is meant by the term punishment? According to the philosopher
Hugo Bedau:
Punishment under law is the authorized imposition of deprivations of freedom or
privacy or other goods to which the person otherwise has a right, or the imposition of special
burdens because the person has been found guilty of some criminal violation, typically
(though not invariably) involving harm to the innocent (Bedau, 2003, italics in original).
Observe the careful use of the terms deprivations and special burdens rather than pain.
In addition, punishment must be for a violation of rules, it must be directed at a
specific offender or offenders, it must be intentionally administered by someone
other than the offender, and it must be unwanted on the part of the offender
(Newman, 1985). However, for punishment to occur, not only should it satisfy
these criteria, but also, according to Beccaria and other scholars, it must have
three qualities to be effective as a deterrent: swiftness, certainty, and severity.
Swiftness (celerity)
Swiftness, or celerity, of punishment involves how closely the punishment follows
the offense, with more immediate punishments thought to have a greater deterrent
effect. Beccaria (1985 [1764], p.55) argued for the promptness of punishment:
The more promptly and the more closely punishment follows upon the commission of a crime the more just and useful it will be. He also contended:
I have said the promptness of punishment is more useful because when the
length of time that passes between the punishment and the misdeed is less, so
much the stronger and more lasting in the human mind is the association of
these two ideas, crime and punishment (Beccaria, 1985 [1764], p.56).
Certainty considers the likelihood of apprehension or the likelihood that punishment attaches for an act. Here, Beccaria argued (1985 [1764], p.58):
One of the greatest curbs on crimes is not the cruelty of punishments, but
their infallibility ... the certainty of a punishment will always make a stronger
impression than the fear of another which is more terrible but combined with
the hope of impunity.

20 Deterrence and delinquency

Moreover, certainty of apprehension and punishment is more important than
either swiftness or severity (see Nagin and Pogarsky, 2001). The wrongdoer must
expect punishment for the behavior.
Beccaria argued that for punishment to be an effective deterrent it must also be
severe. However, when considering this final component of effective punishment,
he was careful to argue that by severity he did not mean that summary executions
for minor violations are an effective deterrent. Rather, by severity Beccaria meant
that the punishment must be proportionate to the harm done to society by the
the purpose can only be to prevent the criminal from inflicting new injuries on its citizens and to deter others from similar acts. Always keeping due
proportions, such punishments and such method of inflicting them ought to
be chosen, therefore, which will make the strongest and most lasting impression on the minds of men, and inflict the least torment on the body of the
criminal. For punishment to attain its end, the evil which it inflicts has only
to exceed the advantage derivable from the crime ... [A]ll beyond this is
superfluous and for that reason tyrannical (Beccaria 1985 [1764], pp.4243).
The end result of deterrence theory is a model that looks roughly like the depiction in Figure 2.1. As portrayed in the diagram, deterrence theory suggests that all
that is necessary to prevent delinquency is punishment that is swift, certain, and
severe. If these qualities are in place, especially certainty, then delinquency will be
held in check.

How to think about deterrence

It is important to realize that deterrence is rather complex. In particular, there
are different ways of thinking about deterrence, different methods to determine
whether deterrence works, and important questions about whether or not juveniles can be deterred from delinquent behavior in the same way that adults can be
deterred from criminal behavior.
What are some types of deterrence?
General deterrence
There are several different ways to categorize deterrence. First, there is a distinction
between general and specific deterrence. General deterrence occurs when adolescents
contemplating wrongdoing opt not to engage in delinquency because they have
seen the consequences of the delinquent act for offenders who have been caught.
Suppose that Bryan contemplates breaking into a car to steal a sound-system.

Deterrence and delinquency21






Figure 2.1 The basic deterrence model.

However, during this period of contemplation, the local police department

announces a crackdown on theft from vehicles, introduces a no-plea policy for
those arrested and charged with larceny from a motor vehicle, and recounts a
story of an adolescent recently arrested for such an offense. Bryan, upon hearing
the announcement, decides to abandon his crime plan; in the language of deterrence theory Bryan has been generally deterred and a crime has been prevented.
Specific deterrence
Specific deterrence focuses not on preventing delinquent acts in general, but rather
on preventing another act from being committed by a particular adolescent. It
is targeted at those individuals already involved in delinquency; the objective of
specific deterrence is to punish the offender with certainty, severity, and swiftness
so that the likelihood of future acts is significantly reduced. Therefore, deterrence
theory can be targeted at the general population or it can be targeted at specific
offenders. The sentencing structure in New York State, for example, allows for
increasing sanctions for repeat offenders even if the offense does not change or the
harm done does not increase. Suppose, for example, that Susan is caught in a buy
and bust operation on the street. She sold a small quantity of crack-cocaine to an
undercover operator. This is her first felony arrest so she pleads guilty and receives
probation as punishment. A few months later she is caught by police a second
time in virtually the same way. However, this time Susan receives a sentence to a
secure juvenile institution. Why the increase in severity of the sanction if the act
committed and the harm done is identical for both offenses? According to specific
deterrence, increasing the sanction the second time is reasonable because obviously the pain of probation was not sufficient to outweigh the pleasure3 derived
from selling crack-cocaine. Therefore, it is necessary to increase the punishment in

22 Deterrence and delinquency

an attempt to deter future drug-dealing behaviors. If, after the period of incarceration, Susan does not go back to selling cocaine then we assume that she has been
specifically deterred from such behavior. On the other hand, what if Susan shifts
from selling crack-cocaine to marijuana or to shoplifting? Has she been deterred?
Absolute and restrictive deterrence
Another issue involves the notion of absolute versus restrictive deterrence (Nagin
and Paternoster, 1991). In general, it is safe to argue that selling crack-cocaine is
a more serious crime than selling marijuana. Suppose that, after a second conviction, Susan shifts away from selling crack-cocaine and instead sells marijuana,
which is far less profitable, but far less serious in terms of the presumed harm it
does to society. Some argue that deterrence has failed in this case as Susan is still
involved in street-level drug sales, but others argue that deterrence has succeeded
because she no longer sells crack-cocaine.
Absolute deterrence focuses on an all or nothing proposition, and only if
Susan completely stops selling drugs is deterrence theory supported. Those who
embrace, though, the idea of restrictive deterrence would probably celebrate
Susans shift in behavior from a more serious to a less serious offense. If it were
not for the punishment for selling crack-cocaine, Susan would continue to sell a
dangerous drug. Therefore, Susan has been deterred. Rather than absolute, the
deterrence has been restrictive.
Does this example illustrate a success or a failure for deterrence theory? There
are no simple answers to this question since people can reasonably disagree on
which is preferable, although most would probably choose absolute deterrence as
the most desirable goal.
Objective and subjective deterrence
Yet another complexity is whether the risk of punishment in terms of certainty,
severity, and swiftness should be objective (what it is in reality) or subjective (what
the individual perceives it to be). Much of the research on deterrence theory
during the mid-twentieth century focused on objective deterrence (Nagin [1998]
provides a review of this research); however, most contemporary research considers what people perceive as risky versus what is actually risky. For example, what
is more risky in terms of likelihood of death: flying in an airplane to Miami or
driving a car to Miami? Actually, driving is far more dangerous (objective risk),
but many people perceive flying to be more dangerous (subjective risk). So, when
we consider risk of punishment from the justice system should we consider what
juveniles perceive the risk of punishment to be, what the actual risk of punishment
is, or both? Similarly, there are different views of what is pleasurable or rewarding.

Research on deterrence and delinquency

Three research strategies have been implemented to ascertain the effects of deterrence on adolescents. The first strategy focuses on research that uses pre-test

Deterrence and delinquency23

intervention post-test studies. In this type of study there is a pre-intervention
assessment of juvenile delinquency, an intervention, and then a post-intervention
assessment to ascertain what proportion of study subjects has stopped offending
and was presumably deterred. The second strategy is to assess whether or not
deterrence works for adolescents the same way it is supposed to work for adults.
This strategy considers the effects, if any, of subjective deterrence on a sample
of adolescents to see if their perceptions of risk predict their involvement in
delinquency. In these studies, deterrence theory is thought to be supported if, for
example, a high perceived risk of apprehension is accompanied by less delinquent
behavior. The final strategy is to compare juvenile justice and criminal justice processing. The criminal justice system is generally considered more punitive than the
juvenile justice system; thus, if deterrence operates as expected one would see a
substantial drop-off in delinquent behavior as the jurisdiction shifts from juvenile
to criminal (general deterrence) and, among those enmeshed in the system, one
would expect to see lower rates of repeat offending for those in the criminal justice
system than those who remain in the juvenile justice system (Levitt, 1998).
Pre-test intervention post-test
Although there are several examples of this type of research, we highlight only
two studies: Scared Straight! and the Pulling Levers Program. Scared Straight! was a program that began in New Jersey and became quite popular. Many jurisdictions
around the United States and in other nations implemented Scared Straight! programs with the goal of deterring petty juvenile offenders from a life of crime.
There are several good websites with Scared Straight!-like themes, and there are also
several documentaries based on the original Scared Straight! program at Rahway
State Prison (now called East Jersey State Prison).
The Scared Straight! program takes at-risk adolescents who have had contact
with the juvenile justice system and sends them on a field trip to a maximum
security prison to meet with a group of inmates doing 25 years to life for violent
crimes. During their field trip the adolescents are processed into the prison by
gruff guards, walk through the cell blocks where they are taunted by inmates
behind bars, and spend time with a group of lifers. The lifers tell them about
life on the inside; they swear, they yell, they intimidate. The developers of Scared
Straight! envisioned a program so forceful, so compelling, and so frightening that
the young offenders would be deterred from subsequent offending. Several documentaries about the program claimed that it was an unqualified success. The most
recent, Scared Straight! 20 Years Later (1999), revisits the original participants in the
1978 documentary to see how they are doing. All are living conventional lives,
with the exception of a few who developed into serious, chronic offenders. Based
upon these accounts, it is difficult to believe that Scared Straight! is anything but an
impressive success that needs to be widely implemented.
However, Finckenauer (1982; Finckenauer and Gavin, 1999) and Petrosino and
colleagues (2000) have argued that scientific evidence does not support the claim
that Scared Straight! is an effective program. In an exhaustive review of research
addressing these programs, Petrosino and colleagues (2000) discovered that not